Driver.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "Ji7JLeMWAKaR"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import cv2\n",
    "from tensorflow.keras.layers import Input, Conv2D, BatchNormalization, Dropout, Flatten, Dense, MaxPool2D\n",
    "from tensorflow.keras.models import Model, Sequential\n",
    "from tensorflow.keras.initializers import glorot_uniform\n",
    "from tensorflow.keras.optimizers import Adam, SGD\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import classification_report\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from sklearn.model_selection import train_test_split\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.metrics import classification_report, confusion_matrix\n",
    "import seaborn as sns\n",
    "from tensorflow.keras.regularizers import l2\n",
    "import tensorflow as tf\n",
    "import pandas as pd\n",
    "from imutils.face_utils import FaceAligner\n",
    "from imutils.face_utils import rect_to_bb\n",
    "import imutils\n",
    "import dlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "glIc2iJRANNL",
    "outputId": "91d55e95-9da5-47eb-db0d-11ac5bdfd3e4"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--2023-08-10 20:03:27--  https://cainvas-static.s3.amazonaws.com/media/user_data/Yuvnish17/driver_drowsiness_detection_modified.zip\n",
      "Resolving cainvas-static.s3.amazonaws.com (cainvas-static.s3.amazonaws.com)... 52.219.156.195, 52.219.160.103, 52.219.158.111, ...\n",
      "Connecting to cainvas-static.s3.amazonaws.com (cainvas-static.s3.amazonaws.com)|52.219.156.195|:443... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 127436131 (122M) [application/x-zip-compressed]\n",
      "Saving to: ‘driver_drowsiness_detection_modified.zip’\n",
      "\n",
      "driver_drowsiness_d 100%[===================>] 121.53M  12.7MB/s    in 11s     \n",
      "\n",
      "2023-08-10 20:03:40 (10.6 MB/s) - ‘driver_drowsiness_detection_modified.zip’ saved [127436131/127436131]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "!wget https://cainvas-static.s3.amazonaws.com/media/user_data/Yuvnish17/driver_drowsiness_detection_modified.zip\n",
    "!unzip -qo driver_drowsiness_detection_modified.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "oVyD8QrOBr-H",
    "outputId": "7e7ea4d8-d712-4baa-a057-5613f62a9a2f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1452\n",
      "1452\n"
     ]
    }
   ],
   "source": [
    "data = np.load('driver_drowsiness_detection/dataset_compressed.npz', allow_pickle=True)\n",
    "X = data['arr_0']\n",
    "Y = data['arr_1']\n",
    "\n",
    "X = list(X)\n",
    "Y = list(Y)\n",
    "print(len(X))\n",
    "print(len(Y))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "6S_yQfhCDJkU",
    "outputId": "b3cbc857-b7f3-4eaf-d77e-cda933805dfa"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1452\n",
      "(32, 32, 3)\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(X)):\n",
    "    img = X[i]\n",
    "    img = cv2.resize(img, (32, 32))\n",
    "    X[i] = img\n",
    "\n",
    "print(len(X))\n",
    "print(X[0].shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "9i9bhWo5DQLB",
    "outputId": "08870cf4-c134-4d40-825d-e1d18851980d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1452,)\n",
      "0\n",
      "{0, 1}\n"
     ]
    }
   ],
   "source": [
    "label_encoder = LabelEncoder()\n",
    "Y = label_encoder.fit_transform(Y)\n",
    "print(Y.shape)\n",
    "print(Y[0])\n",
    "print(set(Y))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "UCCcsrK6DUtS",
    "outputId": "95170739-4820-4416-972b-96a35f9072e0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1452, 32, 32, 3)\n",
      "(1452,)\n"
     ]
    }
   ],
   "source": [
    "X = np.array(X)\n",
    "Y = np.array(Y)\n",
    "print(X.shape)\n",
    "print(Y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "mb1RE42EDZnn",
    "outputId": "c5b918af-2362-4ec1-8970-b09e024c12b4"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1161, 32, 32, 3)\n",
      "(1161,)\n",
      "(291, 32, 32, 3)\n",
      "(291,)\n",
      "(1161, 32, 32, 3)\n",
      "(1161, 2)\n",
      "(291, 32, 32, 3)\n",
      "(291, 2)\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.2)\n",
    "print(X_train.shape)\n",
    "print(Y_train.shape)\n",
    "print(X_test.shape)\n",
    "print(Y_test.shape)\n",
    "\n",
    "Y_train = to_categorical(Y_train)\n",
    "Y_test = to_categorical(Y_test)\n",
    "\n",
    "print(X_train.shape)\n",
    "print(Y_train.shape)\n",
    "print(X_test.shape)\n",
    "print(Y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 871
    },
    "id": "fCaFsclkDirz",
    "outputId": "5fd8d29a-30f9-45bd-f2c8-e55cb4e5dbe6"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure1 = plt.figure(figsize=(5, 5))\n",
    "idx_closed = np.where(Y==0)\n",
    "img_closed = X[idx_closed[0][0]]\n",
    "plt.imshow(img_closed)\n",
    "plt.title('Image of Closed Eye representing Driver is sleeping')\n",
    "plt.axis('off')\n",
    "plt.show()\n",
    "\n",
    "figure2 = plt.figure(figsize=(5, 5))\n",
    "idx_open = np.where(Y==1)\n",
    "img_open = X[idx_open[0][0]]\n",
    "plt.imshow(img_open)\n",
    "plt.title('Image of Open Eye representing Driver is not sleeping')\n",
    "plt.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "D-IsczYMDp48"
   },
   "outputs": [],
   "source": [
    "def driver_drowsiness_detection_model(input_shape=(32, 32, 3)):\n",
    "    model = Sequential()\n",
    "    model.add(Input(shape=input_shape))\n",
    "    model.add(Conv2D(32, (3, 3), padding='same', strides=(1, 1), name='conv1', activation='relu',\n",
    "                     kernel_initializer=glorot_uniform(seed=0)))\n",
    "    model.add(BatchNormalization())\n",
    "    model.add(Conv2D(32, (3, 3), padding='same', strides=(1, 1), name='conv2', activation='relu',\n",
    "                     kernel_initializer=glorot_uniform(seed=0)))\n",
    "    model.add(BatchNormalization())\n",
    "    model.add(Dropout(0.2))\n",
    "    model.add(MaxPool2D((2, 2), strides=(2, 2)))\n",
    "\n",
    "    model.add(Conv2D(64, (3, 3), padding='same', strides=(1, 1), name='conv3', activation='relu',\n",
    "                     kernel_initializer=glorot_uniform(seed=0)))\n",
    "    model.add(BatchNormalization())\n",
    "    model.add(MaxPool2D((2, 2), strides=(2, 2)))\n",
    "    model.add(Conv2D(64, (3, 3), padding='same', strides=(1, 1), name='conv4', activation='relu',\n",
    "                     kernel_initializer=glorot_uniform(seed=0)))\n",
    "    model.add(BatchNormalization())\n",
    "    model.add(Dropout(0.3))\n",
    "    model.add(MaxPool2D((2, 2), strides=(2, 2)))\n",
    "\n",
    "    model.add(Conv2D(64, (3, 3), padding='same', strides=(1, 1), name='conv5', activation='relu',\n",
    "                     kernel_initializer=glorot_uniform(seed=0)))\n",
    "    model.add(BatchNormalization())\n",
    "    model.add(Conv2D(64, (3, 3), padding='same', strides=(1, 1), name='conv6', activation='relu',\n",
    "                     kernel_initializer=glorot_uniform(seed=0)))\n",
    "    model.add(BatchNormalization())\n",
    "    model.add(Conv2D(64, (3, 3), padding='same', strides=(1, 1), name='conv7', activation='relu',\n",
    "                     kernel_initializer=glorot_uniform(seed=0)))\n",
    "    model.add(BatchNormalization())\n",
    "    model.add(Dropout(0.4))\n",
    "    model.add(MaxPool2D((2, 2), strides=(2, 2)))\n",
    "\n",
    "\n",
    "    model.add(Flatten())\n",
    "    model.add(Dense(128, activation='relu', kernel_initializer=glorot_uniform(seed=0), name='fc1'))\n",
    "    model.add(Dropout(0.5))\n",
    "    model.add(Dense(128, activation='relu', kernel_initializer=glorot_uniform(seed=0), name='fc2'))\n",
    "    model.add(Dropout(0.5))\n",
    "    model.add(Dense(2, activation='softmax', kernel_initializer=glorot_uniform(seed=0), name='fc3'))\n",
    "\n",
    "    optimizer = Adam(0.0001)\n",
    "    model.compile(optimizer=optimizer, loss='categorical_crossentropy', metrics=['accuracy'])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "CLGGt7nADwQI",
    "outputId": "3295e942-59af-4dce-ffdc-a69ba794aebb"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv1 (Conv2D)              (None, 32, 32, 32)        896       \n",
      "                                                                 \n",
      " batch_normalization (BatchN  (None, 32, 32, 32)       128       \n",
      " ormalization)                                                   \n",
      "                                                                 \n",
      " conv2 (Conv2D)              (None, 32, 32, 32)        9248      \n",
      "                                                                 \n",
      " batch_normalization_1 (Batc  (None, 32, 32, 32)       128       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 32, 32, 32)        0         \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2D  (None, 16, 16, 32)       0         \n",
      " )                                                               \n",
      "                                                                 \n",
      " conv3 (Conv2D)              (None, 16, 16, 64)        18496     \n",
      "                                                                 \n",
      " batch_normalization_2 (Batc  (None, 16, 16, 64)       256       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " max_pooling2d_1 (MaxPooling  (None, 8, 8, 64)         0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " conv4 (Conv2D)              (None, 8, 8, 64)          36928     \n",
      "                                                                 \n",
      " batch_normalization_3 (Batc  (None, 8, 8, 64)         256       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 8, 8, 64)          0         \n",
      "                                                                 \n",
      " max_pooling2d_2 (MaxPooling  (None, 4, 4, 64)         0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " conv5 (Conv2D)              (None, 4, 4, 64)          36928     \n",
      "                                                                 \n",
      " batch_normalization_4 (Batc  (None, 4, 4, 64)         256       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " conv6 (Conv2D)              (None, 4, 4, 64)          36928     \n",
      "                                                                 \n",
      " batch_normalization_5 (Batc  (None, 4, 4, 64)         256       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " conv7 (Conv2D)              (None, 4, 4, 64)          36928     \n",
      "                                                                 \n",
      " batch_normalization_6 (Batc  (None, 4, 4, 64)         256       \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " dropout_2 (Dropout)         (None, 4, 4, 64)          0         \n",
      "                                                                 \n",
      " max_pooling2d_3 (MaxPooling  (None, 2, 2, 64)         0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 256)               0         \n",
      "                                                                 \n",
      " fc1 (Dense)                 (None, 128)               32896     \n",
      "                                                                 \n",
      " dropout_3 (Dropout)         (None, 128)               0         \n",
      "                                                                 \n",
      " fc2 (Dense)                 (None, 128)               16512     \n",
      "                                                                 \n",
      " dropout_4 (Dropout)         (None, 128)               0         \n",
      "                                                                 \n",
      " fc3 (Dense)                 (None, 2)                 258       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 227,554\n",
      "Trainable params: 226,786\n",
      "Non-trainable params: 768\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model= driver_drowsiness_detection_model(input_shape=(32, 32, 3))\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "wqCVWUy5EIik",
    "outputId": "6f5da0ad-d18f-4623-ce18-11274fc174c3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/200\n",
      "10/10 [==============================] - 13s 762ms/step - loss: 1.3292 - accuracy: 0.5340 - val_loss: 0.7021 - val_accuracy: 0.5155\n",
      "Epoch 2/200\n",
      "10/10 [==============================] - 10s 930ms/step - loss: 1.2778 - accuracy: 0.5366 - val_loss: 0.6772 - val_accuracy: 0.5223\n",
      "Epoch 3/200\n",
      "10/10 [==============================] - 9s 873ms/step - loss: 1.0632 - accuracy: 0.5556 - val_loss: 0.6410 - val_accuracy: 0.5876\n",
      "Epoch 4/200\n",
      "10/10 [==============================] - 8s 792ms/step - loss: 0.9618 - accuracy: 0.5952 - val_loss: 0.6219 - val_accuracy: 0.6632\n",
      "Epoch 5/200\n",
      "10/10 [==============================] - 7s 678ms/step - loss: 0.9031 - accuracy: 0.6064 - val_loss: 0.6168 - val_accuracy: 0.7113\n",
      "Epoch 6/200\n",
      "10/10 [==============================] - 8s 862ms/step - loss: 0.8544 - accuracy: 0.6115 - val_loss: 0.6145 - val_accuracy: 0.7182\n",
      "Epoch 7/200\n",
      "10/10 [==============================] - 7s 688ms/step - loss: 0.7992 - accuracy: 0.6357 - val_loss: 0.6088 - val_accuracy: 0.7079\n",
      "Epoch 8/200\n",
      "10/10 [==============================] - 8s 863ms/step - loss: 0.7829 - accuracy: 0.6107 - val_loss: 0.6016 - val_accuracy: 0.6907\n",
      "Epoch 9/200\n",
      "10/10 [==============================] - 8s 830ms/step - loss: 0.7022 - accuracy: 0.6486 - val_loss: 0.5968 - val_accuracy: 0.6907\n",
      "Epoch 10/200\n",
      "10/10 [==============================] - 7s 696ms/step - loss: 0.7274 - accuracy: 0.6563 - val_loss: 0.5910 - val_accuracy: 0.6804\n",
      "Epoch 11/200\n",
      "10/10 [==============================] - 7s 679ms/step - loss: 0.6993 - accuracy: 0.6606 - val_loss: 0.5815 - val_accuracy: 0.6907\n",
      "Epoch 12/200\n",
      "10/10 [==============================] - 8s 832ms/step - loss: 0.6824 - accuracy: 0.6761 - val_loss: 0.5717 - val_accuracy: 0.6907\n",
      "Epoch 13/200\n",
      "10/10 [==============================] - 8s 828ms/step - loss: 0.6396 - accuracy: 0.6736 - val_loss: 0.5624 - val_accuracy: 0.6976\n",
      "Epoch 14/200\n",
      "10/10 [==============================] - 7s 678ms/step - loss: 0.7332 - accuracy: 0.6581 - val_loss: 0.5486 - val_accuracy: 0.7285\n",
      "Epoch 15/200\n",
      "10/10 [==============================] - 7s 678ms/step - loss: 0.6469 - accuracy: 0.6744 - val_loss: 0.5324 - val_accuracy: 0.7595\n",
      "Epoch 16/200\n",
      "10/10 [==============================] - 7s 672ms/step - loss: 0.6391 - accuracy: 0.6994 - val_loss: 0.5207 - val_accuracy: 0.7663\n",
      "Epoch 17/200\n",
      "10/10 [==============================] - 8s 779ms/step - loss: 0.6167 - accuracy: 0.6960 - val_loss: 0.5097 - val_accuracy: 0.7698\n",
      "Epoch 18/200\n",
      "10/10 [==============================] - 8s 814ms/step - loss: 0.5989 - accuracy: 0.7071 - val_loss: 0.4970 - val_accuracy: 0.7698\n",
      "Epoch 19/200\n",
      "10/10 [==============================] - 7s 662ms/step - loss: 0.5875 - accuracy: 0.7046 - val_loss: 0.4850 - val_accuracy: 0.7869\n",
      "Epoch 20/200\n",
      "10/10 [==============================] - 9s 853ms/step - loss: 0.5675 - accuracy: 0.7304 - val_loss: 0.4732 - val_accuracy: 0.7973\n",
      "Epoch 21/200\n",
      "10/10 [==============================] - 8s 881ms/step - loss: 0.5660 - accuracy: 0.7364 - val_loss: 0.4592 - val_accuracy: 0.8144\n",
      "Epoch 22/200\n",
      "10/10 [==============================] - 7s 653ms/step - loss: 0.5526 - accuracy: 0.7459 - val_loss: 0.4467 - val_accuracy: 0.8213\n",
      "Epoch 23/200\n",
      "10/10 [==============================] - 8s 806ms/step - loss: 0.5167 - accuracy: 0.7433 - val_loss: 0.4354 - val_accuracy: 0.8247\n",
      "Epoch 24/200\n",
      "10/10 [==============================] - 8s 814ms/step - loss: 0.5565 - accuracy: 0.7450 - val_loss: 0.4197 - val_accuracy: 0.8282\n",
      "Epoch 25/200\n",
      "10/10 [==============================] - 7s 697ms/step - loss: 0.5286 - accuracy: 0.7580 - val_loss: 0.4011 - val_accuracy: 0.8247\n",
      "Epoch 26/200\n",
      "10/10 [==============================] - 7s 661ms/step - loss: 0.5191 - accuracy: 0.7554 - val_loss: 0.3856 - val_accuracy: 0.8316\n",
      "Epoch 27/200\n",
      "10/10 [==============================] - 7s 659ms/step - loss: 0.5063 - accuracy: 0.7717 - val_loss: 0.3716 - val_accuracy: 0.8247\n",
      "Epoch 28/200\n",
      "10/10 [==============================] - 7s 653ms/step - loss: 0.5363 - accuracy: 0.7778 - val_loss: 0.3576 - val_accuracy: 0.8351\n",
      "Epoch 29/200\n",
      "10/10 [==============================] - 8s 801ms/step - loss: 0.4597 - accuracy: 0.7881 - val_loss: 0.3424 - val_accuracy: 0.8385\n",
      "Epoch 30/200\n",
      "10/10 [==============================] - 8s 881ms/step - loss: 0.4501 - accuracy: 0.7916 - val_loss: 0.3262 - val_accuracy: 0.8660\n",
      "Epoch 31/200\n",
      "10/10 [==============================] - 7s 692ms/step - loss: 0.4975 - accuracy: 0.7890 - val_loss: 0.3117 - val_accuracy: 0.8729\n",
      "Epoch 32/200\n",
      "10/10 [==============================] - 8s 703ms/step - loss: 0.4556 - accuracy: 0.8079 - val_loss: 0.3000 - val_accuracy: 0.8694\n",
      "Epoch 33/200\n",
      "10/10 [==============================] - 8s 812ms/step - loss: 0.4472 - accuracy: 0.7933 - val_loss: 0.2907 - val_accuracy: 0.8763\n",
      "Epoch 34/200\n",
      "10/10 [==============================] - 7s 673ms/step - loss: 0.4286 - accuracy: 0.8165 - val_loss: 0.2809 - val_accuracy: 0.8694\n",
      "Epoch 35/200\n",
      "10/10 [==============================] - 7s 668ms/step - loss: 0.4222 - accuracy: 0.8105 - val_loss: 0.2700 - val_accuracy: 0.8694\n",
      "Epoch 36/200\n",
      "10/10 [==============================] - 8s 812ms/step - loss: 0.4407 - accuracy: 0.8217 - val_loss: 0.2581 - val_accuracy: 0.8832\n",
      "Epoch 37/200\n",
      "10/10 [==============================] - 7s 731ms/step - loss: 0.3862 - accuracy: 0.8320 - val_loss: 0.2457 - val_accuracy: 0.8866\n",
      "Epoch 38/200\n",
      "10/10 [==============================] - 7s 662ms/step - loss: 0.3848 - accuracy: 0.8295 - val_loss: 0.2347 - val_accuracy: 0.8935\n",
      "Epoch 39/200\n",
      "10/10 [==============================] - 8s 809ms/step - loss: 0.3828 - accuracy: 0.8519 - val_loss: 0.2262 - val_accuracy: 0.9003\n",
      "Epoch 40/200\n",
      "10/10 [==============================] - 8s 834ms/step - loss: 0.3529 - accuracy: 0.8510 - val_loss: 0.2173 - val_accuracy: 0.9038\n",
      "Epoch 41/200\n",
      "10/10 [==============================] - 9s 793ms/step - loss: 0.3248 - accuracy: 0.8613 - val_loss: 0.2117 - val_accuracy: 0.9072\n",
      "Epoch 42/200\n",
      "10/10 [==============================] - 8s 807ms/step - loss: 0.3183 - accuracy: 0.8665 - val_loss: 0.2106 - val_accuracy: 0.8969\n",
      "Epoch 43/200\n",
      "10/10 [==============================] - 7s 662ms/step - loss: 0.3324 - accuracy: 0.8630 - val_loss: 0.2057 - val_accuracy: 0.9003\n",
      "Epoch 44/200\n",
      "10/10 [==============================] - 8s 816ms/step - loss: 0.3417 - accuracy: 0.8630 - val_loss: 0.1967 - val_accuracy: 0.9107\n",
      "Epoch 45/200\n",
      "10/10 [==============================] - 8s 758ms/step - loss: 0.3394 - accuracy: 0.8665 - val_loss: 0.1872 - val_accuracy: 0.9210\n",
      "Epoch 46/200\n",
      "10/10 [==============================] - 7s 654ms/step - loss: 0.3066 - accuracy: 0.8803 - val_loss: 0.1779 - val_accuracy: 0.9278\n",
      "Epoch 47/200\n",
      "10/10 [==============================] - 8s 805ms/step - loss: 0.3101 - accuracy: 0.8742 - val_loss: 0.1702 - val_accuracy: 0.9244\n",
      "Epoch 48/200\n",
      "10/10 [==============================] - 7s 663ms/step - loss: 0.2902 - accuracy: 0.8794 - val_loss: 0.1641 - val_accuracy: 0.9244\n",
      "Epoch 49/200\n",
      "10/10 [==============================] - 7s 662ms/step - loss: 0.2861 - accuracy: 0.9001 - val_loss: 0.1546 - val_accuracy: 0.9347\n",
      "Epoch 50/200\n",
      "10/10 [==============================] - 8s 819ms/step - loss: 0.2773 - accuracy: 0.8889 - val_loss: 0.1504 - val_accuracy: 0.9347\n",
      "Epoch 51/200\n",
      "10/10 [==============================] - 7s 739ms/step - loss: 0.2818 - accuracy: 0.8906 - val_loss: 0.1483 - val_accuracy: 0.9381\n",
      "Epoch 52/200\n",
      "10/10 [==============================] - 7s 668ms/step - loss: 0.2758 - accuracy: 0.9156 - val_loss: 0.1444 - val_accuracy: 0.9381\n",
      "Epoch 53/200\n",
      "10/10 [==============================] - 8s 812ms/step - loss: 0.2522 - accuracy: 0.9147 - val_loss: 0.1423 - val_accuracy: 0.9381\n",
      "Epoch 54/200\n",
      "10/10 [==============================] - 8s 760ms/step - loss: 0.2680 - accuracy: 0.9018 - val_loss: 0.1405 - val_accuracy: 0.9381\n",
      "Epoch 55/200\n",
      "10/10 [==============================] - 7s 662ms/step - loss: 0.2406 - accuracy: 0.9053 - val_loss: 0.1339 - val_accuracy: 0.9485\n",
      "Epoch 56/200\n",
      "10/10 [==============================] - 8s 811ms/step - loss: 0.2337 - accuracy: 0.9165 - val_loss: 0.1308 - val_accuracy: 0.9519\n",
      "Epoch 57/200\n",
      "10/10 [==============================] - 7s 735ms/step - loss: 0.2224 - accuracy: 0.9199 - val_loss: 0.1291 - val_accuracy: 0.9519\n",
      "Epoch 58/200\n",
      "10/10 [==============================] - 7s 673ms/step - loss: 0.2004 - accuracy: 0.9225 - val_loss: 0.1191 - val_accuracy: 0.9450\n",
      "Epoch 59/200\n",
      "10/10 [==============================] - 8s 816ms/step - loss: 0.2164 - accuracy: 0.9190 - val_loss: 0.1171 - val_accuracy: 0.9485\n",
      "Epoch 60/200\n",
      "10/10 [==============================] - 7s 746ms/step - loss: 0.2337 - accuracy: 0.9053 - val_loss: 0.1160 - val_accuracy: 0.9485\n",
      "Epoch 61/200\n",
      "10/10 [==============================] - 7s 667ms/step - loss: 0.1962 - accuracy: 0.9320 - val_loss: 0.1149 - val_accuracy: 0.9553\n",
      "Epoch 62/200\n",
      "10/10 [==============================] - 10s 971ms/step - loss: 0.1790 - accuracy: 0.9294 - val_loss: 0.1200 - val_accuracy: 0.9485\n",
      "Epoch 63/200\n",
      "10/10 [==============================] - 8s 822ms/step - loss: 0.1751 - accuracy: 0.9320 - val_loss: 0.1135 - val_accuracy: 0.9553\n",
      "Epoch 64/200\n",
      "10/10 [==============================] - 7s 664ms/step - loss: 0.1840 - accuracy: 0.9320 - val_loss: 0.1069 - val_accuracy: 0.9622\n",
      "Epoch 65/200\n",
      "10/10 [==============================] - 8s 813ms/step - loss: 0.2385 - accuracy: 0.9147 - val_loss: 0.1062 - val_accuracy: 0.9691\n",
      "Epoch 66/200\n",
      "10/10 [==============================] - 7s 670ms/step - loss: 0.1581 - accuracy: 0.9440 - val_loss: 0.1048 - val_accuracy: 0.9691\n",
      "Epoch 67/200\n",
      "10/10 [==============================] - 8s 762ms/step - loss: 0.1975 - accuracy: 0.9216 - val_loss: 0.1022 - val_accuracy: 0.9656\n",
      "Epoch 68/200\n",
      "10/10 [==============================] - 8s 845ms/step - loss: 0.1921 - accuracy: 0.9251 - val_loss: 0.0980 - val_accuracy: 0.9622\n",
      "Epoch 69/200\n",
      "10/10 [==============================] - 7s 664ms/step - loss: 0.1850 - accuracy: 0.9328 - val_loss: 0.1004 - val_accuracy: 0.9588\n",
      "Epoch 70/200\n",
      "10/10 [==============================] - 8s 890ms/step - loss: 0.1918 - accuracy: 0.9285 - val_loss: 0.0975 - val_accuracy: 0.9656\n",
      "Epoch 71/200\n",
      "10/10 [==============================] - 8s 797ms/step - loss: 0.1766 - accuracy: 0.9328 - val_loss: 0.0914 - val_accuracy: 0.9691\n",
      "Epoch 72/200\n",
      "10/10 [==============================] - 7s 669ms/step - loss: 0.1585 - accuracy: 0.9406 - val_loss: 0.0881 - val_accuracy: 0.9725\n",
      "Epoch 73/200\n",
      "10/10 [==============================] - 7s 666ms/step - loss: 0.1580 - accuracy: 0.9432 - val_loss: 0.0874 - val_accuracy: 0.9691\n",
      "Epoch 74/200\n",
      "10/10 [==============================] - 8s 811ms/step - loss: 0.1346 - accuracy: 0.9535 - val_loss: 0.0890 - val_accuracy: 0.9691\n",
      "Epoch 75/200\n",
      "10/10 [==============================] - 8s 805ms/step - loss: 0.1674 - accuracy: 0.9406 - val_loss: 0.0952 - val_accuracy: 0.9622\n",
      "Epoch 76/200\n",
      "10/10 [==============================] - 7s 665ms/step - loss: 0.1744 - accuracy: 0.9354 - val_loss: 0.0956 - val_accuracy: 0.9588\n",
      "Epoch 77/200\n",
      "10/10 [==============================] - 7s 666ms/step - loss: 0.1555 - accuracy: 0.9449 - val_loss: 0.0914 - val_accuracy: 0.9622\n",
      "Epoch 78/200\n",
      "10/10 [==============================] - 8s 817ms/step - loss: 0.1470 - accuracy: 0.9449 - val_loss: 0.0903 - val_accuracy: 0.9691\n",
      "Epoch 79/200\n",
      "10/10 [==============================] - 8s 886ms/step - loss: 0.1349 - accuracy: 0.9578 - val_loss: 0.0961 - val_accuracy: 0.9588\n",
      "Epoch 80/200\n",
      "10/10 [==============================] - 7s 667ms/step - loss: 0.1406 - accuracy: 0.9492 - val_loss: 0.0880 - val_accuracy: 0.9691\n",
      "Epoch 81/200\n",
      "10/10 [==============================] - 7s 672ms/step - loss: 0.1673 - accuracy: 0.9500 - val_loss: 0.0863 - val_accuracy: 0.9725\n",
      "Epoch 82/200\n",
      "10/10 [==============================] - 7s 674ms/step - loss: 0.1346 - accuracy: 0.9500 - val_loss: 0.0899 - val_accuracy: 0.9691\n",
      "Epoch 83/200\n",
      "10/10 [==============================] - 9s 1s/step - loss: 0.1403 - accuracy: 0.9414 - val_loss: 0.0856 - val_accuracy: 0.9691\n",
      "Epoch 84/200\n",
      "10/10 [==============================] - 8s 806ms/step - loss: 0.1323 - accuracy: 0.9578 - val_loss: 0.0838 - val_accuracy: 0.9691\n",
      "Epoch 85/200\n",
      "10/10 [==============================] - 7s 672ms/step - loss: 0.1229 - accuracy: 0.9509 - val_loss: 0.0881 - val_accuracy: 0.9691\n",
      "Epoch 86/200\n",
      "10/10 [==============================] - 8s 743ms/step - loss: 0.1376 - accuracy: 0.9552 - val_loss: 0.0896 - val_accuracy: 0.9656\n",
      "Epoch 87/200\n",
      "10/10 [==============================] - 8s 822ms/step - loss: 0.1100 - accuracy: 0.9638 - val_loss: 0.0761 - val_accuracy: 0.9691\n",
      "Epoch 88/200\n",
      "10/10 [==============================] - 8s 794ms/step - loss: 0.1300 - accuracy: 0.9475 - val_loss: 0.0785 - val_accuracy: 0.9622\n",
      "Epoch 89/200\n",
      "10/10 [==============================] - 7s 666ms/step - loss: 0.1032 - accuracy: 0.9612 - val_loss: 0.0816 - val_accuracy: 0.9691\n",
      "Epoch 90/200\n",
      "10/10 [==============================] - 8s 819ms/step - loss: 0.1193 - accuracy: 0.9612 - val_loss: 0.0774 - val_accuracy: 0.9759\n",
      "Epoch 91/200\n",
      "10/10 [==============================] - 8s 813ms/step - loss: 0.1075 - accuracy: 0.9655 - val_loss: 0.0702 - val_accuracy: 0.9759\n",
      "Epoch 92/200\n",
      "10/10 [==============================] - 7s 748ms/step - loss: 0.1026 - accuracy: 0.9612 - val_loss: 0.0822 - val_accuracy: 0.9691\n",
      "Epoch 93/200\n",
      "10/10 [==============================] - 8s 885ms/step - loss: 0.1232 - accuracy: 0.9543 - val_loss: 0.0963 - val_accuracy: 0.9691\n",
      "Epoch 94/200\n",
      "10/10 [==============================] - 7s 698ms/step - loss: 0.1053 - accuracy: 0.9612 - val_loss: 0.0798 - val_accuracy: 0.9725\n",
      "Epoch 95/200\n",
      "10/10 [==============================] - 8s 739ms/step - loss: 0.1019 - accuracy: 0.9604 - val_loss: 0.0795 - val_accuracy: 0.9691\n",
      "Epoch 96/200\n",
      "10/10 [==============================] - 8s 812ms/step - loss: 0.1071 - accuracy: 0.9664 - val_loss: 0.0900 - val_accuracy: 0.9725\n",
      "Epoch 97/200\n",
      "10/10 [==============================] - 8s 880ms/step - loss: 0.1100 - accuracy: 0.9638 - val_loss: 0.0877 - val_accuracy: 0.9691\n",
      "Epoch 98/200\n",
      "10/10 [==============================] - 7s 661ms/step - loss: 0.1176 - accuracy: 0.9664 - val_loss: 0.0709 - val_accuracy: 0.9725\n",
      "Epoch 99/200\n",
      "10/10 [==============================] - 7s 661ms/step - loss: 0.1097 - accuracy: 0.9595 - val_loss: 0.0666 - val_accuracy: 0.9759\n",
      "Epoch 100/200\n",
      "10/10 [==============================] - 8s 801ms/step - loss: 0.1125 - accuracy: 0.9621 - val_loss: 0.0778 - val_accuracy: 0.9725\n",
      "Epoch 101/200\n",
      "10/10 [==============================] - 8s 759ms/step - loss: 0.0961 - accuracy: 0.9681 - val_loss: 0.0824 - val_accuracy: 0.9691\n",
      "Epoch 102/200\n",
      "10/10 [==============================] - 7s 667ms/step - loss: 0.0927 - accuracy: 0.9664 - val_loss: 0.0740 - val_accuracy: 0.9622\n",
      "Epoch 103/200\n",
      "10/10 [==============================] - 9s 947ms/step - loss: 0.0860 - accuracy: 0.9664 - val_loss: 0.0766 - val_accuracy: 0.9656\n",
      "Epoch 104/200\n",
      "10/10 [==============================] - 8s 811ms/step - loss: 0.0894 - accuracy: 0.9673 - val_loss: 0.0825 - val_accuracy: 0.9656\n",
      "Epoch 105/200\n",
      "10/10 [==============================] - 7s 668ms/step - loss: 0.1132 - accuracy: 0.9587 - val_loss: 0.0815 - val_accuracy: 0.9725\n",
      "Epoch 106/200\n",
      "10/10 [==============================] - 7s 672ms/step - loss: 0.1055 - accuracy: 0.9638 - val_loss: 0.0626 - val_accuracy: 0.9759\n",
      "Epoch 107/200\n",
      "10/10 [==============================] - 8s 810ms/step - loss: 0.0836 - accuracy: 0.9690 - val_loss: 0.0683 - val_accuracy: 0.9691\n",
      "Epoch 108/200\n",
      "10/10 [==============================] - 8s 823ms/step - loss: 0.0778 - accuracy: 0.9742 - val_loss: 0.0717 - val_accuracy: 0.9691\n",
      "Epoch 109/200\n",
      "10/10 [==============================] - 7s 739ms/step - loss: 0.0957 - accuracy: 0.9673 - val_loss: 0.0717 - val_accuracy: 0.9725\n",
      "Epoch 110/200\n",
      "10/10 [==============================] - 8s 712ms/step - loss: 0.0888 - accuracy: 0.9733 - val_loss: 0.0580 - val_accuracy: 0.9656\n",
      "Epoch 111/200\n",
      "10/10 [==============================] - 8s 814ms/step - loss: 0.0735 - accuracy: 0.9733 - val_loss: 0.0651 - val_accuracy: 0.9759\n",
      "Epoch 112/200\n",
      "10/10 [==============================] - 8s 815ms/step - loss: 0.0745 - accuracy: 0.9655 - val_loss: 0.0490 - val_accuracy: 0.9725\n",
      "Epoch 113/200\n",
      "10/10 [==============================] - 7s 670ms/step - loss: 0.1038 - accuracy: 0.9621 - val_loss: 0.0521 - val_accuracy: 0.9691\n",
      "Epoch 114/200\n",
      "10/10 [==============================] - 7s 665ms/step - loss: 0.1035 - accuracy: 0.9621 - val_loss: 0.0764 - val_accuracy: 0.9759\n",
      "Epoch 115/200\n",
      "10/10 [==============================] - 7s 668ms/step - loss: 0.0949 - accuracy: 0.9612 - val_loss: 0.0672 - val_accuracy: 0.9759\n",
      "Epoch 116/200\n",
      "10/10 [==============================] - 8s 740ms/step - loss: 0.0733 - accuracy: 0.9759 - val_loss: 0.0532 - val_accuracy: 0.9759\n",
      "Epoch 117/200\n",
      "10/10 [==============================] - 8s 818ms/step - loss: 0.0887 - accuracy: 0.9724 - val_loss: 0.0413 - val_accuracy: 0.9794\n",
      "Epoch 118/200\n",
      "10/10 [==============================] - 8s 812ms/step - loss: 0.0612 - accuracy: 0.9750 - val_loss: 0.0500 - val_accuracy: 0.9759\n",
      "Epoch 119/200\n",
      "10/10 [==============================] - 7s 662ms/step - loss: 0.0746 - accuracy: 0.9707 - val_loss: 0.0720 - val_accuracy: 0.9759\n",
      "Epoch 120/200\n",
      "10/10 [==============================] - 7s 779ms/step - loss: 0.0810 - accuracy: 0.9716 - val_loss: 0.0614 - val_accuracy: 0.9759\n",
      "Epoch 121/200\n",
      "10/10 [==============================] - 8s 818ms/step - loss: 0.0867 - accuracy: 0.9690 - val_loss: 0.0567 - val_accuracy: 0.9691\n",
      "Epoch 122/200\n",
      "10/10 [==============================] - 7s 720ms/step - loss: 0.0720 - accuracy: 0.9767 - val_loss: 0.0589 - val_accuracy: 0.9725\n",
      "Epoch 123/200\n",
      "10/10 [==============================] - 9s 819ms/step - loss: 0.0718 - accuracy: 0.9776 - val_loss: 0.0666 - val_accuracy: 0.9759\n",
      "Epoch 124/200\n",
      "10/10 [==============================] - 8s 817ms/step - loss: 0.0777 - accuracy: 0.9742 - val_loss: 0.0627 - val_accuracy: 0.9794\n",
      "Epoch 125/200\n",
      "10/10 [==============================] - 7s 674ms/step - loss: 0.0641 - accuracy: 0.9793 - val_loss: 0.0557 - val_accuracy: 0.9794\n",
      "Epoch 126/200\n",
      "10/10 [==============================] - 7s 661ms/step - loss: 0.0794 - accuracy: 0.9724 - val_loss: 0.0448 - val_accuracy: 0.9828\n",
      "Epoch 127/200\n",
      "10/10 [==============================] - 8s 817ms/step - loss: 0.0729 - accuracy: 0.9690 - val_loss: 0.0494 - val_accuracy: 0.9828\n",
      "Epoch 128/200\n",
      "10/10 [==============================] - 7s 668ms/step - loss: 0.0654 - accuracy: 0.9707 - val_loss: 0.0578 - val_accuracy: 0.9759\n",
      "Epoch 129/200\n",
      "10/10 [==============================] - 7s 670ms/step - loss: 0.0460 - accuracy: 0.9845 - val_loss: 0.0625 - val_accuracy: 0.9759\n",
      "Epoch 130/200\n",
      "10/10 [==============================] - 8s 818ms/step - loss: 0.0579 - accuracy: 0.9776 - val_loss: 0.0560 - val_accuracy: 0.9794\n",
      "Epoch 131/200\n",
      "10/10 [==============================] - 7s 665ms/step - loss: 0.0653 - accuracy: 0.9767 - val_loss: 0.0662 - val_accuracy: 0.9794\n",
      "Epoch 132/200\n",
      "10/10 [==============================] - 8s 813ms/step - loss: 0.0700 - accuracy: 0.9716 - val_loss: 0.1005 - val_accuracy: 0.9656\n",
      "Epoch 133/200\n",
      "10/10 [==============================] - 8s 768ms/step - loss: 0.0954 - accuracy: 0.9742 - val_loss: 0.0893 - val_accuracy: 0.9691\n",
      "Epoch 134/200\n",
      "10/10 [==============================] - 7s 663ms/step - loss: 0.0642 - accuracy: 0.9785 - val_loss: 0.0692 - val_accuracy: 0.9794\n",
      "Epoch 135/200\n",
      "10/10 [==============================] - 8s 781ms/step - loss: 0.0703 - accuracy: 0.9742 - val_loss: 0.0615 - val_accuracy: 0.9828\n",
      "Epoch 136/200\n",
      "10/10 [==============================] - 8s 813ms/step - loss: 0.0626 - accuracy: 0.9767 - val_loss: 0.0628 - val_accuracy: 0.9828\n",
      "Epoch 137/200\n",
      "10/10 [==============================] - 8s 789ms/step - loss: 0.0556 - accuracy: 0.9828 - val_loss: 0.0712 - val_accuracy: 0.9759\n",
      "Epoch 138/200\n",
      "10/10 [==============================] - 7s 667ms/step - loss: 0.0548 - accuracy: 0.9819 - val_loss: 0.0522 - val_accuracy: 0.9759\n",
      "Epoch 139/200\n",
      "10/10 [==============================] - 8s 788ms/step - loss: 0.0896 - accuracy: 0.9655 - val_loss: 0.0562 - val_accuracy: 0.9794\n",
      "Epoch 140/200\n",
      "10/10 [==============================] - 8s 806ms/step - loss: 0.0644 - accuracy: 0.9767 - val_loss: 0.0681 - val_accuracy: 0.9691\n",
      "Epoch 141/200\n",
      "10/10 [==============================] - 7s 671ms/step - loss: 0.0556 - accuracy: 0.9819 - val_loss: 0.0669 - val_accuracy: 0.9725\n",
      "Epoch 142/200\n",
      "10/10 [==============================] - 7s 678ms/step - loss: 0.0742 - accuracy: 0.9785 - val_loss: 0.0633 - val_accuracy: 0.9759\n",
      "Epoch 143/200\n",
      "10/10 [==============================] - 8s 748ms/step - loss: 0.0876 - accuracy: 0.9707 - val_loss: 0.0720 - val_accuracy: 0.9725\n",
      "Epoch 144/200\n",
      "10/10 [==============================] - 8s 706ms/step - loss: 0.0562 - accuracy: 0.9802 - val_loss: 0.0701 - val_accuracy: 0.9725\n",
      "Epoch 145/200\n",
      "10/10 [==============================] - 8s 811ms/step - loss: 0.0710 - accuracy: 0.9759 - val_loss: 0.0612 - val_accuracy: 0.9794\n",
      "Epoch 146/200\n",
      "10/10 [==============================] - 7s 664ms/step - loss: 0.0968 - accuracy: 0.9673 - val_loss: 0.0512 - val_accuracy: 0.9759\n",
      "Epoch 147/200\n",
      "10/10 [==============================] - 7s 675ms/step - loss: 0.0748 - accuracy: 0.9785 - val_loss: 0.0442 - val_accuracy: 0.9828\n",
      "Epoch 148/200\n",
      "10/10 [==============================] - 7s 663ms/step - loss: 0.0542 - accuracy: 0.9811 - val_loss: 0.0435 - val_accuracy: 0.9828\n",
      "Epoch 149/200\n",
      "10/10 [==============================] - 7s 669ms/step - loss: 0.0573 - accuracy: 0.9785 - val_loss: 0.0433 - val_accuracy: 0.9828\n",
      "Epoch 150/200\n",
      "10/10 [==============================] - 8s 814ms/step - loss: 0.0437 - accuracy: 0.9836 - val_loss: 0.0439 - val_accuracy: 0.9828\n",
      "Epoch 151/200\n",
      "10/10 [==============================] - 7s 698ms/step - loss: 0.0602 - accuracy: 0.9811 - val_loss: 0.0545 - val_accuracy: 0.9725\n",
      "Epoch 152/200\n",
      "10/10 [==============================] - 8s 705ms/step - loss: 0.0522 - accuracy: 0.9836 - val_loss: 0.0492 - val_accuracy: 0.9828\n",
      "Epoch 153/200\n",
      "10/10 [==============================] - 8s 810ms/step - loss: 0.0661 - accuracy: 0.9802 - val_loss: 0.0400 - val_accuracy: 0.9794\n",
      "Epoch 154/200\n",
      "10/10 [==============================] - 7s 675ms/step - loss: 0.0521 - accuracy: 0.9785 - val_loss: 0.0355 - val_accuracy: 0.9794\n",
      "Epoch 155/200\n",
      "10/10 [==============================] - 8s 816ms/step - loss: 0.0579 - accuracy: 0.9767 - val_loss: 0.0535 - val_accuracy: 0.9828\n",
      "Epoch 156/200\n",
      "10/10 [==============================] - 8s 812ms/step - loss: 0.0471 - accuracy: 0.9888 - val_loss: 0.0590 - val_accuracy: 0.9794\n",
      "Epoch 157/200\n",
      "10/10 [==============================] - 7s 674ms/step - loss: 0.0667 - accuracy: 0.9759 - val_loss: 0.0728 - val_accuracy: 0.9759\n",
      "Epoch 158/200\n",
      "10/10 [==============================] - 8s 807ms/step - loss: 0.0492 - accuracy: 0.9819 - val_loss: 0.0448 - val_accuracy: 0.9794\n",
      "Epoch 159/200\n",
      "10/10 [==============================] - 8s 816ms/step - loss: 0.0360 - accuracy: 0.9879 - val_loss: 0.0393 - val_accuracy: 0.9828\n",
      "Epoch 160/200\n",
      "10/10 [==============================] - 7s 665ms/step - loss: 0.0525 - accuracy: 0.9828 - val_loss: 0.0459 - val_accuracy: 0.9759\n",
      "Epoch 161/200\n",
      "10/10 [==============================] - 8s 893ms/step - loss: 0.0287 - accuracy: 0.9897 - val_loss: 0.0504 - val_accuracy: 0.9759\n",
      "Epoch 162/200\n",
      "10/10 [==============================] - 8s 759ms/step - loss: 0.0500 - accuracy: 0.9811 - val_loss: 0.0449 - val_accuracy: 0.9759\n",
      "Epoch 163/200\n",
      "10/10 [==============================] - 9s 812ms/step - loss: 0.0365 - accuracy: 0.9897 - val_loss: 0.0495 - val_accuracy: 0.9759\n",
      "Epoch 164/200\n",
      "10/10 [==============================] - 8s 813ms/step - loss: 0.0394 - accuracy: 0.9862 - val_loss: 0.0473 - val_accuracy: 0.9794\n",
      "Epoch 165/200\n",
      "10/10 [==============================] - 8s 813ms/step - loss: 0.0426 - accuracy: 0.9836 - val_loss: 0.0488 - val_accuracy: 0.9863\n",
      "Epoch 166/200\n",
      "10/10 [==============================] - 7s 719ms/step - loss: 0.0440 - accuracy: 0.9854 - val_loss: 0.0547 - val_accuracy: 0.9828\n",
      "Epoch 167/200\n",
      "10/10 [==============================] - 7s 673ms/step - loss: 0.0442 - accuracy: 0.9871 - val_loss: 0.0731 - val_accuracy: 0.9759\n",
      "Epoch 168/200\n",
      "10/10 [==============================] - 8s 818ms/step - loss: 0.0577 - accuracy: 0.9776 - val_loss: 0.0798 - val_accuracy: 0.9725\n",
      "Epoch 169/200\n",
      "10/10 [==============================] - 7s 666ms/step - loss: 0.0355 - accuracy: 0.9871 - val_loss: 0.0673 - val_accuracy: 0.9725\n",
      "Epoch 170/200\n",
      "10/10 [==============================] - 8s 817ms/step - loss: 0.0598 - accuracy: 0.9828 - val_loss: 0.0646 - val_accuracy: 0.9759\n",
      "Epoch 171/200\n",
      "10/10 [==============================] - 7s 665ms/step - loss: 0.0364 - accuracy: 0.9879 - val_loss: 0.0605 - val_accuracy: 0.9794\n",
      "Epoch 172/200\n",
      "10/10 [==============================] - 8s 756ms/step - loss: 0.0362 - accuracy: 0.9871 - val_loss: 0.0572 - val_accuracy: 0.9759\n",
      "Epoch 173/200\n",
      "10/10 [==============================] - 8s 815ms/step - loss: 0.0367 - accuracy: 0.9871 - val_loss: 0.0481 - val_accuracy: 0.9828\n",
      "Epoch 174/200\n",
      "10/10 [==============================] - 7s 668ms/step - loss: 0.0362 - accuracy: 0.9888 - val_loss: 0.0498 - val_accuracy: 0.9759\n",
      "Epoch 175/200\n",
      "10/10 [==============================] - 8s 812ms/step - loss: 0.0460 - accuracy: 0.9862 - val_loss: 0.0564 - val_accuracy: 0.9759\n",
      "Epoch 176/200\n",
      "10/10 [==============================] - 8s 808ms/step - loss: 0.0301 - accuracy: 0.9879 - val_loss: 0.0563 - val_accuracy: 0.9759\n",
      "Epoch 177/200\n",
      "10/10 [==============================] - 7s 670ms/step - loss: 0.0358 - accuracy: 0.9862 - val_loss: 0.0550 - val_accuracy: 0.9725\n",
      "Epoch 178/200\n",
      "10/10 [==============================] - 8s 804ms/step - loss: 0.0328 - accuracy: 0.9888 - val_loss: 0.0559 - val_accuracy: 0.9691\n",
      "Epoch 179/200\n",
      "10/10 [==============================] - 7s 674ms/step - loss: 0.0356 - accuracy: 0.9845 - val_loss: 0.0459 - val_accuracy: 0.9794\n",
      "Epoch 180/200\n",
      "10/10 [==============================] - 7s 676ms/step - loss: 0.0393 - accuracy: 0.9871 - val_loss: 0.0392 - val_accuracy: 0.9794\n",
      "Epoch 181/200\n",
      "10/10 [==============================] - 8s 812ms/step - loss: 0.0368 - accuracy: 0.9854 - val_loss: 0.0600 - val_accuracy: 0.9759\n",
      "Epoch 182/200\n",
      "10/10 [==============================] - 7s 665ms/step - loss: 0.0310 - accuracy: 0.9888 - val_loss: 0.0535 - val_accuracy: 0.9759\n",
      "Epoch 183/200\n",
      "10/10 [==============================] - 7s 668ms/step - loss: 0.0365 - accuracy: 0.9879 - val_loss: 0.0414 - val_accuracy: 0.9828\n",
      "Epoch 184/200\n",
      "10/10 [==============================] - 8s 889ms/step - loss: 0.0275 - accuracy: 0.9914 - val_loss: 0.0518 - val_accuracy: 0.9794\n",
      "Epoch 185/200\n",
      "10/10 [==============================] - 8s 884ms/step - loss: 0.0313 - accuracy: 0.9922 - val_loss: 0.0505 - val_accuracy: 0.9828\n",
      "Epoch 186/200\n",
      "10/10 [==============================] - 7s 667ms/step - loss: 0.0228 - accuracy: 0.9940 - val_loss: 0.0427 - val_accuracy: 0.9863\n",
      "Epoch 187/200\n",
      "10/10 [==============================] - 8s 752ms/step - loss: 0.0393 - accuracy: 0.9879 - val_loss: 0.0458 - val_accuracy: 0.9794\n",
      "Epoch 188/200\n",
      "10/10 [==============================] - 8s 811ms/step - loss: 0.0364 - accuracy: 0.9897 - val_loss: 0.0441 - val_accuracy: 0.9759\n",
      "Epoch 189/200\n",
      "10/10 [==============================] - 8s 784ms/step - loss: 0.0377 - accuracy: 0.9879 - val_loss: 0.0523 - val_accuracy: 0.9759\n",
      "Epoch 190/200\n",
      "10/10 [==============================] - 7s 674ms/step - loss: 0.0384 - accuracy: 0.9879 - val_loss: 0.0431 - val_accuracy: 0.9794\n",
      "Epoch 191/200\n",
      "10/10 [==============================] - 8s 814ms/step - loss: 0.0392 - accuracy: 0.9871 - val_loss: 0.0758 - val_accuracy: 0.9691\n",
      "Epoch 192/200\n",
      "10/10 [==============================] - 8s 765ms/step - loss: 0.0510 - accuracy: 0.9845 - val_loss: 0.0372 - val_accuracy: 0.9794\n",
      "Epoch 193/200\n",
      "10/10 [==============================] - 7s 663ms/step - loss: 0.0459 - accuracy: 0.9802 - val_loss: 0.0405 - val_accuracy: 0.9828\n",
      "Epoch 194/200\n",
      "10/10 [==============================] - 7s 674ms/step - loss: 0.0472 - accuracy: 0.9862 - val_loss: 0.0838 - val_accuracy: 0.9759\n",
      "Epoch 195/200\n",
      "10/10 [==============================] - 7s 802ms/step - loss: 0.0320 - accuracy: 0.9888 - val_loss: 0.0695 - val_accuracy: 0.9759\n",
      "Epoch 196/200\n",
      "10/10 [==============================] - 8s 815ms/step - loss: 0.0474 - accuracy: 0.9845 - val_loss: 0.0517 - val_accuracy: 0.9759\n",
      "Epoch 197/200\n",
      "10/10 [==============================] - 7s 669ms/step - loss: 0.0413 - accuracy: 0.9888 - val_loss: 0.0502 - val_accuracy: 0.9725\n",
      "Epoch 198/200\n",
      "10/10 [==============================] - 7s 667ms/step - loss: 0.0356 - accuracy: 0.9905 - val_loss: 0.0756 - val_accuracy: 0.9725\n",
      "Epoch 199/200\n",
      "10/10 [==============================] - 8s 815ms/step - loss: 0.0340 - accuracy: 0.9879 - val_loss: 0.1006 - val_accuracy: 0.9588\n",
      "Epoch 200/200\n",
      "10/10 [==============================] - 8s 771ms/step - loss: 0.0412 - accuracy: 0.9879 - val_loss: 0.0789 - val_accuracy: 0.9725\n"
     ]
    }
   ],
   "source": [
    "aug = ImageDataGenerator(rotation_range=20, zoom_range=0.2, horizontal_flip=True)\n",
    "hist = model.fit(aug.flow(X_train, Y_train, batch_size=128), batch_size=128, epochs=200, validation_data=(X_test, Y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "sJkXJTTuEWJq",
    "outputId": "3e6de922-b230-4e8d-b3ad-a508e9cd11a1"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure = plt.figure(figsize=(10, 10))\n",
    "plt.plot(hist.history['accuracy'], label='Train_accuracy')\n",
    "plt.plot(hist.history['val_accuracy'], label='Test_accuracy')\n",
    "plt.title('Model Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend(loc=\"upper left\")\n",
    "plt.show()\n",
    "\n",
    "figure2 = plt.figure(figsize=(10, 10))\n",
    "plt.plot(hist.history['loss'], label='Train_loss')\n",
    "plt.plot(hist.history['val_loss'], label='Test_loss')\n",
    "plt.title('Model Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend(loc=\"upper left\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "4qDyOQK9LjQM",
    "outputId": "2bc3633a-8c41-4e65-f21d-62f7c9222939"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10/10 [==============================] - 1s 47ms/step - loss: 0.0789 - accuracy: 0.9725\n",
      "Test Set Accuracy: 0.9725086092948914\n",
      "Test Set Loss: 0.07887206226587296\n"
     ]
    }
   ],
   "source": [
    "pred = model.evaluate(X_test, Y_test)\n",
    "print(f'Test Set Accuracy: {pred[1]}')\n",
    "print(f'Test Set Loss: {pred[0]}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "FatFMYxkLpSl",
    "outputId": "699e432c-c872-4a0e-b514-5a92f1d480ac"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10/10 [==============================] - 1s 46ms/step\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.97      0.97      0.97       141\n",
      "           1       0.97      0.97      0.97       150\n",
      "\n",
      "    accuracy                           0.97       291\n",
      "   macro avg       0.97      0.97      0.97       291\n",
      "weighted avg       0.97      0.97      0.97       291\n",
      "\n"
     ]
    }
   ],
   "source": [
    "ypred = model.predict(X_test)\n",
    "ypred = np.argmax(ypred, axis=1)\n",
    "Y_test_pred = np.argmax(Y_test, axis=1)\n",
    "print(classification_report(Y_test_pred, ypred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 462
    },
    "id": "Q5sYWPdnLsRg",
    "outputId": "d12d2cab-5f95-4ac9-e033-f56020f03a25"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "matrix = confusion_matrix(Y_test_pred, ypred)\n",
    "df_cm = pd.DataFrame(matrix, index=[0, 1], columns=[0, 1])\n",
    "figure = plt.figure(figsize=(5, 5))\n",
    "sns.heatmap(df_cm, annot=True, fmt='d')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "id": "deoWQCkhLxOX"
   },
   "outputs": [],
   "source": [
    "model.save('Driver_Drowsiness_Detection.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "J1n9mjgNL0z4",
    "outputId": "15ddc038-fa50-421f-d7fc-4845c2a7a764"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 30ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n"
     ]
    }
   ],
   "source": [
    "labels = ['Closed', 'Open']\n",
    "img_closed1 = cv2.imread('driver_drowsiness_detection/closed_eye.jpg')\n",
    "img_closed2 = cv2.imread('driver_drowsiness_detection/closed_eye2.jpg')\n",
    "img_open1 = cv2.imread('driver_drowsiness_detection/open_eye.jpg')\n",
    "img_open2 = cv2.imread('driver_drowsiness_detection/open_eye2.jpg')\n",
    "\n",
    "img_closed1 = cv2.resize(img_closed1, (32, 32))\n",
    "img_closed2 = cv2.resize(img_closed2, (32, 32))\n",
    "img_open1 = cv2.resize(img_open1, (32, 32))\n",
    "img_open2 = cv2.resize(img_open2, (32, 32))\n",
    "\n",
    "img_closed1 = np.array(img_closed1)\n",
    "img_closed2 = np.array(img_closed2)\n",
    "img_open1 = np.array(img_open1)\n",
    "img_open2 = np.array(img_open2)\n",
    "\n",
    "img_closed1 = np.expand_dims(img_closed1, axis=0)\n",
    "img_closed2 = np.expand_dims(img_closed2, axis=0)\n",
    "img_open1 = np.expand_dims(img_open1, axis=0)\n",
    "img_open2 = np.expand_dims(img_open2, axis=0)\n",
    "ypred_closed1 = model.predict(img_closed1)\n",
    "ypred_closed2 = model.predict(img_closed2)\n",
    "ypred_open1 = model.predict(img_open1)\n",
    "ypred_open2 = model.predict(img_open2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 213
    },
    "id": "bughz344L5Sn",
    "outputId": "ee3a6352-6945-459c-ecbc-79befb417229"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 200x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure = plt.figure(figsize=(2, 2))\n",
    "img_closed1 = np.squeeze(img_closed1, axis=0)\n",
    "plt.imshow(img_closed1)\n",
    "plt.axis('off')\n",
    "plt.title(f'Prediction by the model: {labels[np.argmax(ypred_closed1[0], axis=0)]}')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 213
    },
    "id": "pYpOHwkoL-5K",
    "outputId": "d7fe3b82-ef36-4880-b502-04ae9c5bc954"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 200x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure = plt.figure(figsize=(2, 2))\n",
    "img_open1 = np.squeeze(img_open1, axis=0)\n",
    "plt.imshow(img_open1)\n",
    "plt.axis('off')\n",
    "plt.title(f'Prediction by the model: {labels[np.argmax(ypred_open1[0], axis=0)]}')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "id": "NIj4yWfAME4-"
   },
   "outputs": [],
   "source": [
    "def full_face_detection_pipeline(input_image_path):\n",
    "    face_cascade = cv2.CascadeClassifier('driver_drowsiness_detection/haarcascade_frontalface_default.xml')\n",
    "    eye_cascade = cv2.CascadeClassifier('driver_drowsiness_detection/haarcascade_eye.xml')\n",
    "    detector = dlib.get_frontal_face_detector()\n",
    "    predictor = dlib.shape_predictor('driver_drowsiness_detection/shape_predictor_68_face_landmarks.dat')\n",
    "    fa = FaceAligner(predictor, desiredFaceWidth=256)\n",
    "    test_image = cv2.imread(input_image_path)\n",
    "    test_image = imutils.resize(test_image, width=800)\n",
    "    test_image_gray = cv2.cvtColor(test_image, cv2.COLOR_BGR2GRAY)\n",
    "    rects = detector(test_image_gray, 2)\n",
    "    for rect in rects:\n",
    "        (x, y, w, h) = rect_to_bb(rect)\n",
    "        faceOrig = imutils.resize(test_image[y:y+h, x:x+w], width=256)\n",
    "        faceAligned = fa.align(test_image, test_image_gray, rect)\n",
    "        faceAligned_gray = cv2.cvtColor(faceAligned, cv2.COLOR_BGR2GRAY)\n",
    "        plt.imshow(faceAligned_gray)\n",
    "        plt.axis('off')\n",
    "        plt.title('Aligned Face')\n",
    "        plt.show()\n",
    "        eyes = eye_cascade.detectMultiScale(faceAligned_gray, 1.1, 4)\n",
    "        predictions = []\n",
    "        for (ex, ey, ew, eh) in eyes:\n",
    "            eye = faceAligned[ey:ey+eh, ex:ex+ew]\n",
    "#             cv2.rectangle(test_image, (x+ex, y+ey), (x+ex+ew, y+ey+eh), (0, 0, 255), 8)\n",
    "            eye = cv2.resize(eye, (32, 32))\n",
    "            eye = np.array(eye)\n",
    "            eye = np.expand_dims(eye, axis=0)\n",
    "            ypred = model.predict(eye)\n",
    "            ypred = np.argmax(ypred[0], axis=0)\n",
    "            predictions.append(ypred)\n",
    "        if all(i==0 for i in predictions):\n",
    "            cv2.rectangle(test_image, (x, y), (x+w, y+h), (0, 0, 255), 8)\n",
    "            cv2.putText(test_image, 'Sleeping', (x, y-10), cv2.FONT_HERSHEY_SIMPLEX, 1.5, (0, 0, 255), 3)\n",
    "        else:\n",
    "            cv2.rectangle(test_image, (x, y), (x+w, y+h), (0, 255, 0), 8)\n",
    "            cv2.putText(test_image, 'Not Sleeping', (x, y-10), cv2.FONT_HERSHEY_SIMPLEX, 1.5, (0, 255, 0), 3)\n",
    "    output_path = 'driver_drowsiness_detection/test_image_prediction.jpg'\n",
    "    cv2.imwrite(output_path, test_image)\n",
    "    return output_path\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "P663zotrMT7K"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}